A trusted data collection method based on lightweight blockchain for the Internet of Things

Abstract

This invention belongs to the interdisciplinary field of blockchain and wireless control networks for the Internet of Things (IoT), specifically relating to a lightweight blockchain-based trusted data collection method for the IoT. To address malicious node attacks in wireless control networks, a lightweight blockchain-based trusted data collection method is proposed. During state observation, this method allows information to be exchanged between adjacent observation nodes. It discovers a series of trusted nodes through a trust model. Based on a designed Proof of Trust (POT) consensus mechanism, all nodes jointly maintain a blockchain for data exchange between a wireless autonomous sensor group and a remote controller, thereby ensuring secure and reliable data collection.

Description

Technical Field

[0001] The present invention belongs to the interdisciplinary technical field of blockchain and wireless control networks of the Internet of Things, and specifically relates to a trusted data collection method based on a lightweight blockchain for the Internet of Things. Background Art

[0002] In wireless control networks for the Industrial Internet of Things (IIoT), sensor data and control commands are exchanged over communication networks for remote data acquisition and distributed control, respectively. System performance depends on the design of data acquisition, transmission, and control. However, due to the harsh environments of industrial sites and the potential for malicious network attacks, the security and reliability of remotely collected data are crucial for accurately estimating the current status of equipment, which in turn affects the effectiveness of remote feedback control.

[0003] When it comes to dynamic state estimation in existing distributed wireless sensor networks, when a network is attacked, the malicious attack quickly spreads throughout the network due to information exchange between nodes, causing significant data errors and even the collapse of the entire system. Previous researchers have proposed various methods to address this issue, but these systems often only provide security for a portion of the system, such as introducing secure nodes, which assumes these nodes are secure. If a malicious attack occurs at a critical node in the system, it will still cause the system to crash, which is unreasonable. Therefore, to better address these issues, the existing distributed architecture has been improved by leveraging the security, tamper-proofing, and transparency of blockchain. Summary of the Invention

[0004] The purpose of this invention is to address the security issues existing in traditional distributed wireless sensor networks and propose a trusted data collection method based on lightweight blockchain.

[0005] This paper first proposes a trust value generation mechanism for sensor nodes in a public blockchain scenario. This mechanism assigns each node in the sensor network an attribute, called a trust value (Tr), which describes the node's reliability at that moment. Based on this node trust value generation mechanism, the present invention proposes a lightweight blockchain consensus algorithm that, unlike traditional consensus algorithms, can quickly and securely reach effective consensus. Finally, the present invention proposes a blockchain-based interaction model between controllers and sensors in a wireless control network, enabling effective coordination between controllers and sensors.

[0006] To facilitate the description of the present invention, the terms and models used in the present invention are first introduced:

[0007] Wireless Communication: A communication method that uses the property of electromagnetic wave signals that can propagate in free space to exchange information.

[0008] Internet of Things: A network formed by combining various information sensing devices with the Internet.

[0009] Blockchain: A chain of blocks containing data.

[0010] Remote Controller (RC): A control entity located at a remote location that issues control commands.

[0011] Wireless Controlled Equipment (WCE): A control object that receives control instructions from a remote controller.

[0012] Wireless Autonomous Sensor (WAS): autonomously observes the status of wireless controlled devices and provides feedback to remote controllers.

[0013] Proof of Trust (PoT): This paper proposes a lightweight consensus mechanism for proof of trust between nodes on a blockchain.

[0014] The technical solution adopted in the present invention is:

[0015] A lightweight blockchain-based trusted data collection method for the Internet of Things, comprising the following steps:

[0016] S1. Define a public chain scenario where each node in the sensor network has a trust value Tr. Tr describes the reliability of the current node at this moment:

[0017]

[0018] where Φ is a continuous random variable modeled by the trust of WAS nodes and follows a Beta distribution with parameters a and b, τ is the penalty factor, and the subscript t refers to the tth update; a and b represent the number of good behaviors and the total number of collections during the data collection period, respectively.

[0019] S2, define the error statistical threshold α, the maximum error threshold β, when the observation error Er of sensor n n <α is considered good behavior, when the observation error α of sensor n <Er n <β is considered as malicious behavior. When the observation error Er of sensor n n>β is considered as intolerable malicious behavior.

[0020] S3. Define the trust weight Trw according to the trust value Tr. For a network containing N MAS nodes, the trust value Tr of the nth MAS node is n Normalized to trust weight:

[0021]

[0022] S4. Data broadcast: When the observation task is in progress, the nodes in the blockchain network broadcast their local state estimates to the entire network.

[0023] S5. Trust broadcast: Nodes within the blockchain network calculate the trust values ​​of their neighboring nodes and broadcast them to the network.

[0024] S6. Data block packaging: Nodes in the blockchain network package multiple state data blocks and trust data blocks into one block and add a block header.

[0025] S7. Competition for block maintenance rights: The node with the highest trust value obtains the block maintenance rights. If there are multiple WAS nodes with the highest current trust values, the trust values ​​recorded in the previous block of each node with the highest current trust value are checked and compared again. If there is a single WAS node with the highest current trust value, then that WAS node obtains the maintenance rights for the current block. Otherwise, the trust values ​​of each node in the previous block are continuously searched until a single node with the highest trust value is found. This node then obtains the maintenance rights for the current block.

[0026] S8. Verification: Nodes within the network verify the signatures of the data blocks in the block to determine whether the newly generated block is valid. If more than 50% of the nodes verify that it is valid, the block will be added to the blockchain. If more than 50% of the nodes verify that it is invalid, the node with the second highest trust value will obtain maintenance rights. If there are multiple nodes with the second highest trust values, the trust values ​​recorded in the previous block of each node with the second highest trust value will be checked and compared again. At this time, if there is a single WAS node with the highest current trust value, then this WAS node obtains maintenance rights for the current block. Otherwise, the trust values ​​of each node in the previous block will be continuously searched until the single node with the highest trust value is found, and this node obtains maintenance rights for the current block.

[0027] The beneficial effect of the present invention is that, in response to malicious node attacks in wireless control networks, a trusted data collection method based on a lightweight blockchain is proposed. When performing state observation, it allows information to be exchanged between adjacent observation nodes, and finds a series of trusted nodes through a trust model. According to the designed PoT consensus mechanism, all nodes jointly maintain a blockchain for data exchange between wireless autonomous sensor groups and remote controllers, thereby ensuring the safe and reliable collection of data. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 System model diagram of a distributed wireless control network interacting through blockchain.

[0029] Figure 2 It is the MAS node trust change curve.

[0030] Figure 3 This is a comparison curve of the control effects between the traditional method and the trust consensus method. DETAILED DESCRIPTION

[0031] The technical solution of the present invention is described in detail below with reference to examples and drawings:

[0032] like Figure 1 Figure 2 shows a schematic diagram of a distributed wireless control network interacting via blockchain in the present invention. The RC sends different control commands to multiple WCEs in the system via blockchain. Simultaneously, the WAS observes and samples the status of the WCEs and also feeds back the current status of the WCEs to the RC via blockchain. Blockchain in the system actually improves the wireless control network's anti-attack capabilities, enhancing the security and reliability of the entire wireless sensor network.

[0033] Consider a common discrete state equation of a feedback control system with AWGN (Gaussian white noise) as follows

[0034] X t =AX t-1 +Bu t-1 +W t-1

[0035] X t Represents the state vector of the system at time t, A and B are the system state transfer matrix and control matrix respectively, u t is the control command at time t, W t is Gaussian white noise at time t. Six MAS nodes are set to serve one MCE node, and two of them are randomly selected to be attacked by malicious attacks. Possible attack methods include increasing observation noise and random additive interference on the observation data.

[0036] According to the existing noise environment, the present invention sets the penalty factor τ = 2 and the error statistical threshold α = 5. When the observation error is less than 5, the MAS node is counted as a good behavior. On the contrary, if the observation error is greater than 5, the MAS node is counted as a malicious behavior. The maximum error threshold β = 15 is set. When the single observation error of the MAS node exceeds 15, the trust value Tr will be set to 0. At the same time, let a t =u t ,b t =τ(nt -u t ).

[0037] Each MAS uses distributed Kalman filtering to make local state estimation for WCE. In the Mth trust update cycle, the local estimation value of the nth MAS at time t is defined as For the first MAS n , using the trust value Tr obtained in the (N-1)th trust update cycle N-1 , weight the local estimates given by each MAS except itself in this trust update cycle to obtain a weighted average estimate As the estimated reference value of the observation error statistics, for a certain MAS, its own estimated error value is not included in the reference calculation to prevent the error of a single MAS from being too large and affecting the overall mean. as follows,

[0038]

[0039] Then MAS at time t n The observation error is

[0040]

[0041] The two thresholds α and β set above are used to deal with the small-scale continuous change interference and large-scale jump interference that may exist in malicious interference. When Er≤β, α is used as the threshold to statistically update u in a trust update cycle. t , n t Value

[0042] When Er≤α

[0043] a t =a t-1 +1 b t =b t-1 +1

[0044] When Er>α

[0045] a t =a t-1 b t =b t-1 +1

[0046] At this time a t , b t It is also dynamically updated as the observation proceeds and is used to calculate the trust value Tr.

[0047] Increase the observation noise at WAS1 to simulate the bad observation environment that the wireless sensor node may encounter in actual work. In this case, the error of the node fluctuates around the error statistical threshold but does not exceed the maximum error threshold. Under this condition, the trust value change curve of each node in the wireless sensor network is simulated as follows: Figure 2 ,The simulation found that the trust value of WAS1 in the figure first decreased, but eventually, ,because the observation noise is essentially additive Gaussian white noise, ,under the condition of constant mean, the trust value of WAS1 node ,will eventually stabilize near the constant value. Figure 3 The comparison curve of the control effect of the traditional control method and the method of the present invention is shown in the figure. It can be seen that when a malicious node attack occurs, the convergence curve of the traditional data collection method shows obvious fluctuation deviation. Compared with the traditional data collection method, the state curve converges better when introducing the trust proof consensus method.

Claims

1. A trusted data collection method based on lightweight blockchain for the Internet of Things, characterized by: The following steps are involved: S1. Define a public chain scenario where each node in the sensor network has a trust value Tr. Tr describes the reliability of the current node at this moment: Where Φ is a continuous random variable modeled by the trust of WAS nodes and follows a Beta distribution with parameters a and b. τ is the penalty factor, and the subscript t refers to the tth update. a and b represent the number of good behaviors and the total number of collections during the data collection period, respectively. S2, define the error statistical threshold α, the maximum error threshold β, when the observation error Er of sensor n n <α is considered good behavior, when the observation error α of sensor n <Er n <β is considered as malicious behavior. When the observation error Er of sensor n n >β is considered as intolerable malicious behavior; S3. Define the trust weight Trw according to the trust value Tr. For a network containing N MAS nodes, the trust value Tr of the nth MAS node is n Normalized to trust weight: S4, data broadcast: When the observation task is in progress, the nodes in the blockchain network broadcast their local state estimates to the entire network; S5. Trust broadcast: Nodes in the blockchain network calculate the trust values ​​of their neighboring nodes and broadcast them to the network; S6. Data block packaging: Nodes in the blockchain network package multiple state data blocks and trust data blocks into one block and add a block header; S7. Competition for block maintenance rights: The node with the highest trust value obtains the block maintenance rights. If there are multiple nodes with the same and highest trust values, and if there are multiple WAS nodes with the highest current trust values, the trust values ​​recorded in the previous block of each node with the highest current trust value are checked and compared again. If there is a unique WAS node with the highest current trust value, then the WAS node obtains the maintenance rights of the current block. Otherwise, the trust values ​​of each node in the previous block are continuously searched until a unique node with the highest trust value is found. This node obtains the maintenance rights of the current block. S8. Verification: Nodes in the network verify the signature of the data block in the block to determine whether the newly generated block is valid. If more than 50% of the nodes verify that it is valid, the block will be added to the blockchain; if more than 50% of the nodes verify that it is invalid, the node with the second highest trust value will obtain the maintenance right. If there are multiple nodes with the second highest trust value, the trust value recorded in the previous block of each node with the second highest trust value will be checked and compared again; at this time, if there is a unique WAS node with the highest current trust value, the WAS node will obtain the maintenance right of the current block, otherwise, continue to search for the trust value of each node in the previous block until the unique node with the highest trust value is found, and the node will obtain the maintenance right of the current block.

Images